““Alignment” between what humans want and what machines do will continue to be a critical, unsolved problem.”
Our reading, the sentence that was checked:
Alignment between human wants and machine behavior will remain a critical unsolved problem.
Assessment: Supported Settled before its deadline on 2026-10-29 Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, What to Expect When You’re Expecting … GPT-5, paragraph 8.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-what-to-expect-when-youre-expecting-62e-a9525d-a1
We recorded 15 claims from this post. See every claim from this post.
The wording above is the version this site observed on 2026-10-28. The source may have been edited after 2025-07-27. This page does not claim the wording was captured then.
What the claim commits to
- Kind
- Forecast
- Deadline, as the speaker wrote it
- None stated. This site does not add one.
- Deadline state when assessed
- Settled before its deadline
- Deadline state on 2026-10-29
- Settled before its deadline (the release’s reference date)
From the speaker
- “Alignment” between what humans want and what machines do
- will continue
- a critical, unsolved problem
Our interpretation
- The surrounding passage makes reliable helpfulness, harmlessness and truthfulness the relevant meaning of alignment.
- The ordinary reading is that material alignment failures will persist into the GPT-5 era, not that alignment can never be solved.
- Critical means consequential for dependable use, not a specified probability of catastrophe.
- No deadline is supplied. A documented post-forecast continuation of material alignment problems can establish the predicted persistence.
Counts as supported if: Post-forecast primary evaluations or disclosures document material failures to behave as intended and identify their mitigation as unresolved.
Counts as contradicted if: Broad, reproducible evidence establishes a general solution that reliably prevents the relevant failures across realistic and adversarial conditions. Improvements alone do not contradict persistence.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Did the forecast come true?
Outcome: Supported
The dated evidence matches what the claim committed to, under the criteria frozen before grading.
Scored on a plain reading. The speaker named no measure for this claim, so it was judged on the reading an informed reader would take, stated below under our interpretation. Both reviews had to find that reading fair.
- Evidence as of
- Deadline state when assessed
- Settled before its deadline
- Assessment recorded
- Policy version
- 2026-10-28.7
Rationale
OpenAI's August 13, 2025 GPT-5 system card identifies deception mitigation as an open research problem and documents incomplete mitigation. That establishes the predicted continuation into the GPT-5 era. Later deployment research provides additional support. Reported improvements are real counterweight to a claim of no progress, but Marcus's sentence says the problem remains unsolved.
Limitations
- This supports persistence, not a prediction that alignment will remain unsolved forever.
- The evidence primarily concerns deception and truthfulness, rather than every interpretation of human values.
Evidence (3 items)
Supports the claim: OpenAI,
open research challenge.
Why it matters here: The dated GPT-5 disclosure explicitly identifies deception mitigation as unresolved. Its surrounding discussion describes incomplete mitigations and residual failures, establishing that this alignment problem persisted into the GPT-5 era.
- Publisher
- OpenAI
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Supports the claim: OpenAI,
deceptive behaviors at deployment.
Why it matters here: The report describes misleading tool-use behavior in GPT-5.1 deployment. This supports the general alignment-persistence claim; it is not evidence about the original GPT-5's standalone capabilities.
- Publisher
- OpenAI
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Context only: OpenAI,
improvements in reducing hallucinations
Why it matters here: OpenAI reports improvements in a relevant dimension of truthfulness. This limits any reading that alignment work made no progress, but does not establish that alignment was solved.
- Publisher
- OpenAI
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, What to Expect When You’re Expecting … GPT-5
“Alignment” between what humans want and what machines do will continue to be a critical, unsolved problem.
- Where
- paragraph 8
- Published
- 2025-07-27
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-what-to-expect-when-youre-expecting-62e-a9525d-a1: Current. Outcome: Supported. Evidence as of 2026-10-28. Public date not recorded.
Recorded events
Assessment publishedBy this site
First assessment published after both reviews and the final-text check passed.
About assessment marcus-what-to-expect-when-youre-expecting-62e-a9525d-a1.